# microsoft/RecAI

Bridging LLM and Recommender System.

Repository: https://github.com/microsoft/RecAI
Canonical: https://ross.abutalabs.com/products/recai
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: llm, recommender
Last push: 2026-01-27T03:54:54+00:00

## Health v2 (maintenance only)
Score: 56/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 64, release rhythm 35, longevity 77
- inputs: {"age_days": 1091, "days_push": 218, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1192, forks 128 (observed 2026-08-28T04:03:56.365573+00:00)

## What it is
RecAI is a Microsoft research project exploring ways to integrate large language models into recommender systems (LLM4Rec). It includes a conversational recommender AI agent, selective knowledge plugins via prompting, and LLM-based model explainers.

## Use cases
- build a conversational recommender system on top of an existing recommendation model
- make matrix factorization recommenders interactive and explainable with an LLM agent
- inject domain knowledge into an LLM via personalized prompting without fine-tuning
- fine-tune a language model to act as a recommender
- evaluate LLM-based recommendation quality
- use an LLM to explain recommendations from a traditional model

## When to choose
- you want to add LLM-driven interactivity or explainability to an existing recommender system
- you are researching LLM4Rec techniques like agents, prompting, or fine-tuning for recommendations
- you want a reference implementation of a recommender AI agent that calls traditional models as tools

## When to avoid
- you need a production-ready, turnkey recommender system without LLM integration
- you want a lightweight library with a stable API rather than research code and notebooks
- your project has no need for conversational or explainable recommendation

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, llm-inference, rag, machine-learning, chatbot
- domain: artificial-intelligence, large-language-models, machine-learning
- platform: python
- tags: llm4rec, recommender-systems, conversational-recommendation, jupyter-notebooks, microsoft-research, ai-agents, retrieval-augmented-generation

## Member repositories
- microsoft/RecAI (main) score 56

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:56.365573+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:22:31.168526+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/RecAI (fetched 2026-08-28T04:03:56.365573+00:00, sha ae5951b2d7ce)
- Data as of 2026-08-30T08:39:29.467469+00:00.